End-stage lung disease (ESLD) affects over 1 million patients in the U.S. and has an estimated annual economic impact exceeding $ 50 billion. Lung transplantation (LT), the only curative therapy for ESLD, improves survival as well as quality of life. Scarcity of donor organs remains the predominant barrier towards wider application of LT. Despite this discrepancy, only 20% of brain dead donors are considered for LT. Two predominant factors account for the low lung utilization rates in donors. Firstly, though the International Society for Heart and Lung Transplantation has proposed guidelines for lung donor assessment, these recommendations are fairly broad and are variably implemented, leading to significant heterogeneity in assessment of organs for potential transplantation. Unfortunately, there are no validated instruments available to inform donor lung utilization, thereby hampering optimization of a limited resource. Secondly, the impact of donor quality on early graft function after LT is unknown. Severe primary graft dysfunction (PGD) after LT leads to significant morbidity and mortality and lowers long-term survival. Donor factors associated with severe PGD remain inadequately understood and lead to further uncertainty in the decision to accept organs. In this proposal we will address both the principal reasons for low lung utilization rates nationally.
Aim 1 : To develop and validate a predictive model for lung utilization in brain dead donors. With the access to large, prospectively maintained database providing detailed donor level information for donors whose lungs were accepted or declined for LT, we will use multivariable analyses to create a predictive model for likelihood of lung utilization from a brain dead donor. A nomogram will be developed and validated in independent cohorts from other organ procurement organizations (OPOs) and presented as an electronic app.
Aim 2 : To understand the impact of donor factors on early outcomes in lung transplant recipients. We will evaluate lung donors at three collaborating OPOs with well-maintained databases. Detailed information on 90-day outcomes in recipients will be obtained from institutional and national registry data. We will develop multivariable models to understand the impact of donor clinical and CT scan imaging characteristics on the risk of early graft dysfunction and 90-day mortality after LT. The models will be externally validated and will be used to generate a donor score that can predict lung performance after transplant. By developing a tool to guide donor selection for LT and by delineating donor characteristics that impact early outcomes in LT recipients, we will address two critical questions for any clinician evaluating a donor offer: Can we accept these lungs? Will they function adequately? Our findings will be easily incorporated into routine donor care and will guide LT clinicians and policymakers in optimal management of a scarce resource. The models developed would also be readily adaptable for other solid organ transplants.
Lung transplantation (LT), the only curative therapy for end stage lung disease, improves survival as well as quality of life. Scarcity of donor organs remains the predominant barrier towards wider application of LT, yet organs from fewer than 25% of donors are used for LT. In this proposal we will use predictive modeling to address the principal factors accounting for low lung utilization rates in organ donors.
